5370A
I just got a 5370A locally and am going through the performance
checks. Things are not as they should be. Is there someone who knows
this unit intimately who would be willing to help me off list?
Joe Gray
KA5ZEC
Joseph Gray wrote:
I just got a 5370A locally and am going through the performance
checks. Things are not as they should be. Is there someone who knows
this unit intimately who would be willing to help me off list?
Joe Gray
KA5ZEC
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
There seem to be a few people here very familiar with these units. I
too have just bought one and have started going through the perf tests.
Mine, however, suffered a rough trip and has DSP14 blank - either the
display is gone or the anode driver.
I'll admit I prefer people help out on the list - no question is too
stupid and if you really don't want to read it just delete the thread.
This group seems to have more 5370 problem solvers than the hpAgilent
group but that is another option for you.
Good luck with you're checks.
Tim
Joseph Gray wrote:
I just got a 5370A locally and am going through the performance
checks. Things are not as they should be. Is there someone who knows
this unit intimately who would be willing to help me off list?
Joe Gray
KA5ZEC
Joe
What do you have in the way of test gear?
A 100+MHz scope would be useful as well as a spectrum analyser.
Bruce
Please don't take this discussion off list. There are lots of us interested
in anything about 5370A/B maintenance. Thanks!
John, K1AE
-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On
Behalf Of Joseph Gray
Sent: Saturday, October 10, 2009 12:47 AM
To: Discussion of precise time and frequency measurement
Subject: [time-nuts] 5370A
I just got a 5370A locally and am going through the performance
checks. Things are not as they should be. Is there someone who knows
this unit intimately who would be willing to help me off list?
Joe Gray
KA5ZEC
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Tim,
Thanks for the encouragement. I hope you get yours fixed as well.
Joe Gray
KA5ZEC
On Fri, Oct 9, 2009 at 11:26 PM, swingbyte swingbyte@exemail.com.au wrote:
Joseph Gray wrote:
I just got a 5370A locally and am going through the performance
checks. Things are not as they should be. Is there someone who knows
this unit intimately who would be willing to help me off list?
Joe Gray
KA5ZEC
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
There seem to be a few people here very familiar with these units. I too
have just bought one and have started going through the perf tests. Mine,
however, suffered a rough trip and has DSP14 blank - either the display is
gone or the anode driver. I'll admit I prefer people help out on the list -
no question is too stupid and if you really don't want to read it just
delete the thread. This group seems to have more 5370 problem solvers than
the hpAgilent group but that is another option for you.
Good luck with you're checks.
Tim
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Bruce,
I have a 100MHz scope, DVM, frequency counter, signal generator, etc.
Sorry, no spec an. I wish I had one.
The first problem is that it fails the first performance test. Instead
of 100nS, I get about 10nS and that slowly increases to about 12nS.
This is with the unit already warmed up and it is repeatable. I
checked the internal reference output and it is very close to 10MHz,
so that isn't the problem.
Before I go any further, I think I'll open it up reseat all the
boards. It couldn't hurt.
Joe
On Fri, Oct 9, 2009 at 11:51 PM, Bruce Griffiths
bruce.griffiths@xtra.co.nz wrote:
Joseph Gray wrote:
I just got a 5370A locally and am going through the performance
checks. Things are not as they should be. Is there someone who knows
this unit intimately who would be willing to help me off list?
Joe Gray
KA5ZEC
Joe
What do you have in the way of test gear?
A 100+MHz scope would be useful as well as a spectrum analyser.
Bruce
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
It looks like I have a bad BNC on the front panel. After inspecting
the insides, I powered it up again. I got 104nS. Wiggling the BNC, it
went back to 12nS. I tried another coax - same problem. Before I tear
the front panel off, I'll see if I can find in the manual what I have
to adjust to get it to read 100nS +/- 1nS.
One odd thing is that sometimes when I power the unit on, the "Remote"
light is on and sometimes not.
Joe Gray
KA5ZEC
On Sat, Oct 10, 2009 at 12:46 AM, Joseph Gray jgray@zianet.com wrote:
Bruce,
I have a 100MHz scope, DVM, frequency counter, signal generator, etc.
Sorry, no spec an. I wish I had one.
The first problem is that it fails the first performance test. Instead
of 100nS, I get about 10nS and that slowly increases to about 12nS.
This is with the unit already warmed up and it is repeatable. I
checked the internal reference output and it is very close to 10MHz,
so that isn't the problem.
Before I go any further, I think I'll open it up reseat all the
boards. It couldn't hurt.
Joe
On Fri, Oct 9, 2009 at 11:51 PM, Bruce Griffiths
bruce.griffiths@xtra.co.nz wrote:
Joseph Gray wrote:
I just got a 5370A locally and am going through the performance
checks. Things are not as they should be. Is there someone who knows
this unit intimately who would be willing to help me off list?
Joe Gray
KA5ZEC
Joe
What do you have in the way of test gear?
A 100+MHz scope would be useful as well as a spectrum analyser.
Bruce
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Joe
Just be careful no to break the plastic handles off the boards.
The board retention force can be quite high.
Bruce
Joseph Gray wrote:
Bruce,
I have a 100MHz scope, DVM, frequency counter, signal generator, etc.
Sorry, no spec an. I wish I had one.
The first problem is that it fails the first performance test. Instead
of 100nS, I get about 10nS and that slowly increases to about 12nS.
This is with the unit already warmed up and it is repeatable. I
checked the internal reference output and it is very close to 10MHz,
so that isn't the problem.
Before I go any further, I think I'll open it up reseat all the
boards. It couldn't hurt.
Joe
On Fri, Oct 9, 2009 at 11:51 PM, Bruce Griffiths
bruce.griffiths@xtra.co.nz wrote:
Joseph Gray wrote:
I just got a 5370A locally and am going through the performance
checks. Things are not as they should be. Is there someone who knows
this unit intimately who would be willing to help me off list?
Joe Gray
KA5ZEC
Joe
What do you have in the way of test gear?
A 100+MHz scope would be useful as well as a spectrum analyser.
Bruce
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Bruce,
Thanks for the tip. I was very careful. Only one board was really
stuck. I left it alone for now.
Just for the heck of it, I did a few more steps in the Chapter 4
Operation Verification. As I said, discounting the bad BNC, step 13
reads aproximately 103.6nS, so I continued.
Steps 14, 15 and 16 are ok.
Step 17 shows 3.6nS (should be < 1nS). Period Compliment shows
-96.3nS. Something very wrong here.
Steps 18, 19 and 20 are ok.
I'm going to stop there for the night and drag out the scope tomorrow
to pickup at step 21.
Thanks for the help so far.
Joe Gray
KA5ZEC
On Sat, Oct 10, 2009 at 1:50 AM, Bruce Griffiths
bruce.griffiths@xtra.co.nz wrote:
Joe
Just be careful no to break the plastic handles off the boards.
The board retention force can be quite high.
Bruce
Joseph Gray wrote:
Bruce,
I have a 100MHz scope, DVM, frequency counter, signal generator, etc.
Sorry, no spec an. I wish I had one.
The first problem is that it fails the first performance test. Instead
of 100nS, I get about 10nS and that slowly increases to about 12nS.
This is with the unit already warmed up and it is repeatable. I
checked the internal reference output and it is very close to 10MHz,
so that isn't the problem.
Before I go any further, I think I'll open it up reseat all the
boards. It couldn't hurt.
Joe
On Fri, Oct 9, 2009 at 11:51 PM, Bruce Griffiths
bruce.griffiths@xtra.co.nz wrote:
Joseph Gray wrote:
I just got a 5370A locally and am going through the performance
checks. Things are not as they should be. Is there someone who knows
this unit intimately who would be willing to help me off list?
Joe Gray
KA5ZEC
Joe
What do you have in the way of test gear?
A 100+MHz scope would be useful as well as a spectrum analyser.
Bruce
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Oh, one last question for the night. I noticed that the CPU board has
an emply ROM socket. I assume this is normal?
OK, let's make it two questions. While I'm working on this thing,
should I pull the EPROMS and ROMS and read them? I've heard of EPROMS
going bad in some other old equipment (maybe it was on this list). Of
course, finding replacement chips might be a challenge.
Joe Gray
KA5ZEC
On Sat, Oct 10, 2009 at 2:11 AM, Joseph Gray jgray@zianet.com wrote:
Bruce,
Thanks for the tip. I was very careful. Only one board was really
stuck. I left it alone for now.
Just for the heck of it, I did a few more steps in the Chapter 4
Operation Verification. As I said, discounting the bad BNC, step 13
reads aproximately 103.6nS, so I continued.
Steps 14, 15 and 16 are ok.
Step 17 shows 3.6nS (should be < 1nS). Period Compliment shows
-96.3nS. Something very wrong here.
Steps 18, 19 and 20 are ok.
I'm going to stop there for the night and drag out the scope tomorrow
to pickup at step 21.
Thanks for the help so far.
Joe Gray
KA5ZEC
On Sat, Oct 10, 2009 at 1:50 AM, Bruce Griffiths
bruce.griffiths@xtra.co.nz wrote:
Joe
Just be careful no to break the plastic handles off the boards.
The board retention force can be quite high.
Bruce
Joseph Gray wrote:
Bruce,
I have a 100MHz scope, DVM, frequency counter, signal generator, etc.
Sorry, no spec an. I wish I had one.
The first problem is that it fails the first performance test. Instead
of 100nS, I get about 10nS and that slowly increases to about 12nS.
This is with the unit already warmed up and it is repeatable. I
checked the internal reference output and it is very close to 10MHz,
so that isn't the problem.
Before I go any further, I think I'll open it up reseat all the
boards. It couldn't hurt.
Joe
On Fri, Oct 9, 2009 at 11:51 PM, Bruce Griffiths
bruce.griffiths@xtra.co.nz wrote:
Joseph Gray wrote:
I just got a 5370A locally and am going through the performance
checks. Things are not as they should be. Is there someone who knows
this unit intimately who would be willing to help me off list?
Joe Gray
KA5ZEC
Joe
What do you have in the way of test gear?
A 100+MHz scope would be useful as well as a spectrum analyser.
Bruce
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Bruce,
Thanks for the tip. I was very careful. Only one board was really
stuck. I left it alone for now.
Just for the heck of it, I did a few more steps in the Chapter 4
Operation Verification. As I said, discounting the bad BNC, step 13
reads aproximately 103.6nS, so I continued.
Steps 14, 15 and 16 are ok.
Step 17 shows 3.6nS (should be < 1nS). Period Compliment shows
-96.3nS. Something very wrong here.
What are your trigger levels? A small departure from 0.00 volts could
account for this.
Also, what's the standard deviation when measuring its own OCXO?
-- john, KE5FX
Oh, one last question for the night. I noticed that the CPU board has
an emply ROM socket. I assume this is normal?
OK, let's make it two questions. While I'm working on this thing,
should I pull the EPROMS and ROMS and read them? I've heard of EPROMS
going bad in some other old equipment (maybe it was on this list). Of
course, finding replacement chips might be a challenge.
It's usually a good idea to back up the firmware in older instruments,
unless it's not socketed or otherwise safely accessible. For the 5370A, the
ROM images are already available -- see
http://www.ko4bb.com/cgi-bin/manuals.pl?dir=04%29_ROM_Images/HP_-_5370A .
It's also been observed that the firmware sockets can be somewhat flaky in
the 5370As, so watch out for trouble there.
-- john, KE5FX
Oh, one last question for the night. I noticed that the CPU board has
an emply ROM socket. I assume this is normal?
Yes, and you can use it to eliminate the ROM board altogether if desired.
See the .txt and .jpg files in the directory I mentioned earlier.
-- john, KE5FX
John,
Trigger levels are set to "Preset" for most steps. In step 18, I
verified that Preset is 0V and the Level voltage varied from
aproximately -1.3 to +0.6 volts. The manual specifies "approximately
-1.3 to +0.5 volts."
In step 15, the standard deviation was bouncing around between about
36-57pS. The manual says that less than 100pS is good.
Time for bed.
Joe Gray
KA5ZEC
On Sat, Oct 10, 2009 at 2:32 AM, John Miles jmiles@pop.net wrote:
Bruce,
Thanks for the tip. I was very careful. Only one board was really
stuck. I left it alone for now.
Just for the heck of it, I did a few more steps in the Chapter 4
Operation Verification. As I said, discounting the bad BNC, step 13
reads aproximately 103.6nS, so I continued.
Steps 14, 15 and 16 are ok.
Step 17 shows 3.6nS (should be < 1nS). Period Compliment shows
-96.3nS. Something very wrong here.
What are your trigger levels? A small departure from 0.00 volts could
account for this.
Also, what's the standard deviation when measuring its own OCXO?
-- john, KE5FX
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
I keep thinking of things instead of going to bed.
In the 5370A manual, the ops verification of the HPIB port requires an
old HP computer and a program on tape. Needless to say, I don't think
I'm going to come across either of these items. So, has anyone managed
to duplicate that test with modern tools? Or, should I not worry about
testing the port if it works at all?
Now I really need to hit the sack. Goodnight (or morning) to all.
Joe Gray
KA5ZEC
On Sat, Oct 10, 2009 at 3:14 AM, Joseph Gray jgray@zianet.com wrote:
John,
Trigger levels are set to "Preset" for most steps. In step 18, I
verified that Preset is 0V and the Level voltage varied from
aproximately -1.3 to +0.6 volts. The manual specifies "approximately
-1.3 to +0.5 volts."
In step 15, the standard deviation was bouncing around between about
36-57pS. The manual says that less than 100pS is good.
Time for bed.
Joe Gray
KA5ZEC
On Sat, Oct 10, 2009 at 2:32 AM, John Miles jmiles@pop.net wrote:
Bruce,
Thanks for the tip. I was very careful. Only one board was really
stuck. I left it alone for now.
Just for the heck of it, I did a few more steps in the Chapter 4
Operation Verification. As I said, discounting the bad BNC, step 13
reads aproximately 103.6nS, so I continued.
Steps 14, 15 and 16 are ok.
Step 17 shows 3.6nS (should be < 1nS). Period Compliment shows
-96.3nS. Something very wrong here.
What are your trigger levels? A small departure from 0.00 volts could
account for this.
Also, what's the standard deviation when measuring its own OCXO?
-- john, KE5FX
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Hi Joe
I have a similar problem so I would be pleased to hear from you with any
useful information. The basic problem with mine is that the power supply is
suspect - it works, but there is excessive heat, much greater than any other
HP instrument that I have, from the rear mounted heat-sink, this is too hot
to hold your hand on ! There is also a very obvious buzz from the power
transformer, even when it is on "stand-by" - so I guess at worst it could be
shorted turns in the power transformer, or probably better, a faulty bridge
rectifier,or one or more of the large reservoir caps - this an early model
(1984), so perhaps it is to be suspected. I also think the performance is
not to specification, but I will check-out the perceived problem with PSU
before I investigate this matter. I am just about to start the
investigation of the prime problem.
Roy
----- Original Message -----
From: "Joseph Gray" jgray@zianet.com
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Saturday, October 10, 2009 5:46 AM
Subject: [time-nuts] 5370A
I just got a 5370A locally and am going through the performance
checks. Things are not as they should be. Is there someone who knows
this unit intimately who would be willing to help me off list?
Joe Gray
KA5ZEC
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Joe
You can check a lot of the functionality from the front panel.
Make sure that it works well using the front panel controls before
checking that the HPIB interface works.
To check the HPIB functionality you will need an HPIB/GPIB interface for
your PC.
It is easy to duplicate the tests done by the program on the 9825 tape
using Ulrich's EZGPIB (if you use windows).
Otherwise if you use Linux or FreeBSD, etc., and have a Prologix
LAN/GPIB interface then you can program the tests using Python.
Bruce
Joseph Gray wrote:
I keep thinking of things instead of going to bed.
In the 5370A manual, the ops verification of the HPIB port requires an
old HP computer and a program on tape. Needless to say, I don't think
I'm going to come across either of these items. So, has anyone managed
to duplicate that test with modern tools? Or, should I not worry about
testing the port if it works at all?
Now I really need to hit the sack. Goodnight (or morning) to all.
Joe Gray
KA5ZEC
On Sat, Oct 10, 2009 at 3:14 AM, Joseph Gray jgray@zianet.com wrote:
John,
Trigger levels are set to "Preset" for most steps. In step 18, I
verified that Preset is 0V and the Level voltage varied from
aproximately -1.3 to +0.6 volts. The manual specifies "approximately
-1.3 to +0.5 volts."
In step 15, the standard deviation was bouncing around between about
36-57pS. The manual says that less than 100pS is good.
Time for bed.
Joe Gray
KA5ZEC
On Sat, Oct 10, 2009 at 2:32 AM, John Miles jmiles@pop.net wrote:
Bruce,
Thanks for the tip. I was very careful. Only one board was really
stuck. I left it alone for now.
Just for the heck of it, I did a few more steps in the Chapter 4
Operation Verification. As I said, discounting the bad BNC, step 13
reads aproximately 103.6nS, so I continued.
Steps 14, 15 and 16 are ok.
Step 17 shows 3.6nS (should be < 1nS). Period Compliment shows
-96.3nS. Something very wrong here.
What are your trigger levels? A small departure from 0.00 volts could
account for this.
Also, what's the standard deviation when measuring its own OCXO?
-- john, KE5FX
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Joe,
Here's a coarse way to use the front panel to see if any large DC bias
errors have developed in your Input Amp and/or Schmitt Trigger circuits (A3
and A4) for either or both channels. (Disclaimer: this doesn't replace the
corresponding DC bias adjustments in the manual, but it is a quick way to
observe the symptoms that get progressively worse when either an Input Amp
or Schmitt Trigger I.C. are beginning to go bad.)
For each input channel (and while watching its "Trigger" LED) manually
rotate its "Level" pot back and forth, putting the pot in the center of the
range in which the "Trigger" LED repetitively blinks. At this pot position,
your displayed "Trigger Level" should be close to zero volts and the knob's
pointer should be pointing approximately straight up (12 o'clock position).
For this test you want just barely enough input signal so that there's a
very narrow, "exact" setting of the "Trigger Level" pot that produces
triggering.
Assuming you're using the 5370A's timebase output as the source - and to
make it easier to "eyeball" the center of trigger range - configure your
inputs for:
Start Com
50 ohm Input
Divide by 10 attenuation
For each channel, if triggering is centered near zero volts (say, +/-
0.03V), that's a good sign. However, if triggering centers at a larger DC
offset, that may be a sign of trouble in an Input Amp and/or Schmitt
Trigger. Small errors may be adjusted-out via their respective DC bias
adjustments. However as the I.C.s continue to deteriorate, there comes a
time when the only solution is to replace the bad I.C.(s).
Hopefully your unit isn't so afflicted. ...Just covering the bases...
Best,
Greg
----- Original Message -----
From: "Joseph Gray" jgray@zianet.com
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Saturday, October 10, 2009 3:14 AM
Subject: Re: [time-nuts] 5370A
John,
Trigger levels are set to "Preset" for most steps. In step 18, I
verified that Preset is 0V and the Level voltage varied from
aproximately -1.3 to +0.6 volts. The manual specifies "approximately
-1.3 to +0.5 volts."
In step 15, the standard deviation was bouncing around between about
36-57pS. The manual says that less than 100pS is good.
Time for bed.
Joe Gray
KA5ZEC
Roy,
The HP 5370A rear mounted heat-sink typically heats up to around 61 degrees
C. This is very hot to the touch, so I suspect your unit's temperature is
normal (unless it's running way hotter than 61 degrees C).
The buzzing transformer might just be due to loose, vibrating laminations.
You might try tightening the transformer's bolts to see if that reduces the
buzz?
Best,
Greg
P.S. I agree that the rear mounted heat sinks of most other HP equipment run
significantly cooler. The models that run so hot to the touch are the
5370A/B, 5359A, and many of the legacy HP pulse generators (from Boblingen
Division).
----- Original Message -----
From: "Roy Phillips" phill.r1@btinternet.com
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Saturday, October 10, 2009 4:08 AM
Subject: Re: [time-nuts] 5370A
Hi Joe
I have a similar problem so I would be pleased to hear from you with any
useful information. The basic problem with mine is that the power supply is
suspect - it works, but there is excessive heat, much greater than any other
HP instrument that I have, from the rear mounted heat-sink, this is too hot
to hold your hand on ! There is also a very obvious buzz from the power
transformer, even when it is on "stand-by" - so I guess at worst it could be
shorted turns in the power transformer, or probably better, a faulty bridge
rectifier,or one or more of the large reservoir caps - this an early model
(1984), so perhaps it is to be suspected. I also think the performance is
not to specification, but I will check-out the perceived problem with PSU
before I investigate this matter. I am just about to start the
investigation of the prime problem.
Roy
One thing to note, the 5370A has a slim line relay (K1) in the power supply
that switches all of the voltages on... it is there for the crystal oscillator.
I had one that didn't produce enough contract pressure to reliably switch on all
of the voltages, and would quit if you bumped it lightly. I had to readjust
the contact spring pressures, and it became reliable.
-Chuck Harris
Greg Burnett wrote:
Roy,
The HP 5370A rear mounted heat-sink typically heats up to around 61 degrees
C. This is very hot to the touch, so I suspect your unit's temperature is
normal (unless it's running way hotter than 61 degrees C).
The buzzing transformer might just be due to loose, vibrating laminations.
You might try tightening the transformer's bolts to see if that reduces the
buzz?
Best,
Greg
P.S. I agree that the rear mounted heat sinks of most other HP equipment run
significantly cooler. The models that run so hot to the touch are the
5370A/B, 5359A, and many of the legacy HP pulse generators (from Boblingen
Division).
Greg
Thank you - that's good news re temperature - I will check the transformer
for loose laminations. I am collecting the various information from other's
regarding calibration/ performance.
Roy
----- Original Message -----
From: "Greg Burnett" gbusg@comcast.net
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Saturday, October 10, 2009 4:24 PM
Subject: Re: [time-nuts] 5370A
Roy,
The HP 5370A rear mounted heat-sink typically heats up to around 61
degrees
C. This is very hot to the touch, so I suspect your unit's temperature is
normal (unless it's running way hotter than 61 degrees C).
The buzzing transformer might just be due to loose, vibrating laminations.
You might try tightening the transformer's bolts to see if that reduces
the
buzz?
Best,
Greg
P.S. I agree that the rear mounted heat sinks of most other HP equipment
run
significantly cooler. The models that run so hot to the touch are the
5370A/B, 5359A, and many of the legacy HP pulse generators (from Boblingen
Division).
----- Original Message -----
From: "Roy Phillips" phill.r1@btinternet.com
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Saturday, October 10, 2009 4:08 AM
Subject: Re: [time-nuts] 5370A
Hi Joe
I have a similar problem so I would be pleased to hear from you with any
useful information. The basic problem with mine is that the power supply
is
suspect - it works, but there is excessive heat, much greater than any
other
HP instrument that I have, from the rear mounted heat-sink, this is too
hot
to hold your hand on ! There is also a very obvious buzz from the power
transformer, even when it is on "stand-by" - so I guess at worst it could
be
shorted turns in the power transformer, or probably better, a faulty
bridge
rectifier,or one or more of the large reservoir caps - this an early
model
(1984), so perhaps it is to be suspected. I also think the performance is
not to specification, but I will check-out the perceived problem with PSU
before I investigate this matter. I am just about to start the
investigation of the prime problem.
Roy
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Greg,
I did your test and at the 12 o'clock position, both levels read
-0.32V. Zero volts is at about 1 o'clock. With the attenuator switched
in, the range on the pots is fairly sharp.
Thanks for the tip. I'll add the DC bias adjustment to my list.
Joe Gray
KA5ZEC
On Sat, Oct 10, 2009 at 8:57 AM, Greg Burnett gbusg@comcast.net wrote:
Joe,
Here's a coarse way to use the front panel to see if any large DC bias
errors have developed in your Input Amp and/or Schmitt Trigger circuits (A3
and A4) for either or both channels. (Disclaimer: this doesn't replace the
corresponding DC bias adjustments in the manual, but it is a quick way to
observe the symptoms that get progressively worse when either an Input Amp
or Schmitt Trigger I.C. are beginning to go bad.)
For each input channel (and while watching its "Trigger" LED) manually
rotate its "Level" pot back and forth, putting the pot in the center of the
range in which the "Trigger" LED repetitively blinks. At this pot position,
your displayed "Trigger Level" should be close to zero volts and the knob's
pointer should be pointing approximately straight up (12 o'clock position).
For this test you want just barely enough input signal so that there's a
very narrow, "exact" setting of the "Trigger Level" pot that produces
triggering.
Assuming you're using the 5370A's timebase output as the source - and to
make it easier to "eyeball" the center of trigger range - configure your
inputs for:
Start Com
50 ohm Input
Divide by 10 attenuation
For each channel, if triggering is centered near zero volts (say, +/-
0.03V), that's a good sign. However, if triggering centers at a larger DC
offset, that may be a sign of trouble in an Input Amp and/or Schmitt
Trigger. Small errors may be adjusted-out via their respective DC bias
adjustments. However as the I.C.s continue to deteriorate, there comes a
time when the only solution is to replace the bad I.C.(s).
Hopefully your unit isn't so afflicted. ...Just covering the bases...
Best,
Greg
----- Original Message -----
From: "Joseph Gray" jgray@zianet.com
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Saturday, October 10, 2009 3:14 AM
Subject: Re: [time-nuts] 5370A
John,
Trigger levels are set to "Preset" for most steps. In step 18, I
verified that Preset is 0V and the Level voltage varied from
aproximately -1.3 to +0.6 volts. The manual specifies "approximately
-1.3 to +0.5 volts."
In step 15, the standard deviation was bouncing around between about
36-57pS. The manual says that less than 100pS is good.
Time for bed.
Joe Gray
KA5ZEC
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
The heatsink on my unit is quite hot, also. I'm thinking about putting
a fan on it after I get everything working.
On Sat, Oct 10, 2009 at 9:24 AM, Greg Burnett gbusg@comcast.net wrote:
Roy,
The HP 5370A rear mounted heat-sink typically heats up to around 61 degrees
C. This is very hot to the touch, so I suspect your unit's temperature is
normal (unless it's running way hotter than 61 degrees C).
The buzzing transformer might just be due to loose, vibrating laminations.
You might try tightening the transformer's bolts to see if that reduces the
buzz?
Best,
Greg
P.S. I agree that the rear mounted heat sinks of most other HP equipment run
significantly cooler. The models that run so hot to the touch are the
5370A/B, 5359A, and many of the legacy HP pulse generators (from Boblingen
Division).
----- Original Message -----
From: "Roy Phillips" phill.r1@btinternet.com
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Saturday, October 10, 2009 4:08 AM
Subject: Re: [time-nuts] 5370A
Hi Joe
I have a similar problem so I would be pleased to hear from you with any
useful information. The basic problem with mine is that the power supply is
suspect - it works, but there is excessive heat, much greater than any other
HP instrument that I have, from the rear mounted heat-sink, this is too hot
to hold your hand on ! There is also a very obvious buzz from the power
transformer, even when it is on "stand-by" - so I guess at worst it could be
shorted turns in the power transformer, or probably better, a faulty bridge
rectifier,or one or more of the large reservoir caps - this an early model
(1984), so perhaps it is to be suspected. I also think the performance is
not to specification, but I will check-out the perceived problem with PSU
before I investigate this matter. I am just about to start the
investigation of the prime problem.
Roy
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Joe - When you manually set the trigger level pots to the center of the
triggered range (as indicated by flashing trigger LED), does your digital
display indicate trigger levels near zero volts? That is what's important -
and what might be way off if you have DC bias problems in the front-end. If
you have that problem, typically it will be in one channel only, although it
could of course be in both.
Greg
----- Original Message -----
From: "Joseph Gray" jgray@zianet.com
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Saturday, October 10, 2009 12:42 PM
Subject: Re: [time-nuts] 5370A
Greg,
I did your test and at the 12 o'clock position, both levels read
-0.32V. Zero volts is at about 1 o'clock. With the attenuator switched
in, the range on the pots is fairly sharp.
Thanks for the tip. I'll add the DC bias adjustment to my list.
Joe Gray
KA5ZEC
Greg,
Yes, the trigger levels are near zero when the LEDs are flashing. The
level pots are both at 1 o'clock, instead of 12 o'clock, however.
There does seem to be a problem with the preset trigger postition.
Although the display shows 0.00V when at preset, I don't think it is.
When I manually set the levels to 0.00V, the TI shows 100nS. At preset
levels, the TI is noticeably higher.
Joe Gray
KA5ZEC
On Sat, Oct 10, 2009 at 2:08 PM, Greg Burnett gbusg@comcast.net wrote:
Joe - When you manually set the trigger level pots to the center of the
triggered range (as indicated by flashing trigger LED), does your digital
display indicate trigger levels near zero volts? That is what's important -
and what might be way off if you have DC bias problems in the front-end. If
you have that problem, typically it will be in one channel only, although it
could of course be in both.
Greg
----- Original Message -----
From: "Joseph Gray" jgray@zianet.com
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Saturday, October 10, 2009 12:42 PM
Subject: Re: [time-nuts] 5370A
Greg,
I did your test and at the 12 o'clock position, both levels read
-0.32V. Zero volts is at about 1 o'clock. With the attenuator switched
in, the range on the pots is fairly sharp.
Thanks for the tip. I'll add the DC bias adjustment to my list.
Joe Gray
KA5ZEC
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Of course, make sure that the existing fan is running and that the top
instrument cover is installed. (The top cover must be in place so that the
internal fan air currents will loop around and pass through vents, including
the through-holes in the rear heat sink.)
I don't think it would be a good idea to run the instrument too long without
its top cover.
Greg
----- Original Message -----
From: "Joseph Gray" jgray@zianet.com
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Saturday, October 10, 2009 1:55 PM
Subject: Re: [time-nuts] 5370A
The heatsink on my unit is quite hot, also. I'm thinking about putting
a fan on it after I get everything working.
Hi All: I put extended fins on my 1980's Hp equipment by drilling and
tapping the existing sink. I used the white heatsink grease from Radio
Shack to help.
Don Latham
Joseph Gray
The heatsink on my unit is quite hot, also. I'm thinking about putting
a fan on it after I get everything working.
On Sat, Oct 10, 2009 at 9:24 AM, Greg Burnett gbusg@comcast.net wrote:
Roy,
The HP 5370A rear mounted heat-sink typically heats up to around 61
degrees
C. This is very hot to the touch, so I suspect your unit's temperature
is
normal (unless it's running way hotter than 61 degrees C).
The buzzing transformer might just be due to loose, vibrating
laminations.
You might try tightening the transformer's bolts to see if that reduces
the
buzz?
Best,
Greg
P.S. I agree that the rear mounted heat sinks of most other HP equipment
run
significantly cooler. The models that run so hot to the touch are the
5370A/B, 5359A, and many of the legacy HP pulse generators (from
Boblingen
Division).
----- Original Message -----
From: "Roy Phillips" phill.r1@btinternet.com
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Saturday, October 10, 2009 4:08 AM
Subject: Re: [time-nuts] 5370A
Hi Joe
I have a similar problem so I would be pleased to hear from you with any
useful information. The basic problem with mine is that the power supply
is
suspect - it works, but there is excessive heat, much greater than any
other
HP instrument that I have, from the rear mounted heat-sink, this is too
hot
to hold your hand on ! There is also a very obvious buzz from the power
transformer, even when it is on "stand-by" - so I guess at worst it
could be
shorted turns in the power transformer, or probably better, a faulty
bridge
rectifier,or one or more of the large reservoir caps - this an early
model
(1984), so perhaps it is to be suspected. I also think the performance
is
not to specification, but I will check-out the perceived problem with
PSU
before I investigate this matter. I am just about to start the
investigation of the prime problem.
Roy
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
--
Dr. Don Latham AJ7LL
Six Mile Systems LLP
17850 Six Mile Road
POB 134
Huson, MT, 59846
VOX 406-626-4304
www.lightningforensics.com
www.sixmilesystems.com
So now all your HP equipment looks like a 1950's parking lot, huh?
On Sat, Oct 10, 2009 at 4:16 PM, Don Latham djl@montana.com wrote:
Hi All: I put extended fins on my 1980's Hp equipment by drilling and
tapping the existing sink. I used the white heatsink grease from Radio
Shack to help.
Don Latham
Joseph Gray
The heatsink on my unit is quite hot, also. I'm thinking about putting
a fan on it after I get everything working.
On Sat, Oct 10, 2009 at 9:24 AM, Greg Burnett gbusg@comcast.net wrote:
Roy,
The HP 5370A rear mounted heat-sink typically heats up to around 61
degrees
C. This is very hot to the touch, so I suspect your unit's temperature
is
normal (unless it's running way hotter than 61 degrees C).
The buzzing transformer might just be due to loose, vibrating
laminations.
You might try tightening the transformer's bolts to see if that reduces
the
buzz?
Best,
Greg
P.S. I agree that the rear mounted heat sinks of most other HP equipment
run
significantly cooler. The models that run so hot to the touch are the
5370A/B, 5359A, and many of the legacy HP pulse generators (from
Boblingen
Division).
----- Original Message -----
From: "Roy Phillips" phill.r1@btinternet.com
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Saturday, October 10, 2009 4:08 AM
Subject: Re: [time-nuts] 5370A
Hi Joe
I have a similar problem so I would be pleased to hear from you with any
useful information. The basic problem with mine is that the power supply
is
suspect - it works, but there is excessive heat, much greater than any
other
HP instrument that I have, from the rear mounted heat-sink, this is too
hot
to hold your hand on ! There is also a very obvious buzz from the power
transformer, even when it is on "stand-by" - so I guess at worst it
could be
shorted turns in the power transformer, or probably better, a faulty
bridge
rectifier,or one or more of the large reservoir caps - this an early
model
(1984), so perhaps it is to be suspected. I also think the performance
is
not to specification, but I will check-out the perceived problem with
PSU
before I investigate this matter. I am just about to start the
investigation of the prime problem.
Roy
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
--
Dr. Don Latham AJ7LL
Six Mile Systems LLP
17850 Six Mile Road
POB 134
Huson, MT, 59846
VOX 406-626-4304
www.lightningforensics.com
www.sixmilesystems.com
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Joe,
I think 1 o'clock (instead of 12 o'clock) might still be OK and/or typical?
Maybe some other 5370A users can tell us where their Trigger Level pots are
pointing when centered for triggering on a weak signal. (The pots point to
12 o'clock on my 5370B, but I can't remember exactly where they might point
on a typical "A".)
I do know that some of the 5370A A3/A4 Input Amp & Schmitt Trigger I.C.s
developed "I.C. rot" many years after manufacture (due to a contamination
problem during manufacture). For units so afflicted, the symptom was that
"Preset" trigger level was no longer correct, but the trigger level could
still be successfully set manually. For these cases, as the years progressed
(and the "rot" worsened), you had to turn the manual trigger level pot
further and further to one direction to find trigger. As the years of
deterioration continued, there came a point where both trigger level
setability and sensitivity were no longer acceptable.
If you have only a small trigger level issue in "Preset" mode, my thinking
is that you can still correct for it by doing the A3/A4 input adjustments.
Best,
Greg
----- Original Message -----
From: "Joseph Gray" jgray@zianet.com
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Saturday, October 10, 2009 2:23 PM
Subject: Re: [time-nuts] 5370A
Greg,
Yes, the trigger levels are near zero when the LEDs are flashing. The
level pots are both at 1 o'clock, instead of 12 o'clock, however.
There does seem to be a problem with the preset trigger postition.
Although the display shows 0.00V when at preset, I don't think it is.
When I manually set the levels to 0.00V, the TI shows 100nS. At preset
levels, the TI is noticeably higher.
Joe Gray
KA5ZEC
Greg,
I plan on doing the A3/A4 adjustments. I just don't know where I'm
going to get a pulse generator or the fancy scope with the 1GHz
sampling plugin. Not the type of equipment on your average test bench.
On Sat, Oct 10, 2009 at 5:03 PM, Greg Burnett gbusg@comcast.net wrote:
Joe,
I think 1 o'clock (instead of 12 o'clock) might still be OK and/or typical?
Maybe some other 5370A users can tell us where their Trigger Level pots are
pointing when centered for triggering on a weak signal. (The pots point to
12 o'clock on my 5370B, but I can't remember exactly where they might point
on a typical "A".)
I do know that some of the 5370A A3/A4 Input Amp & Schmitt Trigger I.C.s
developed "I.C. rot" many years after manufacture (due to a contamination
problem during manufacture). For units so afflicted, the symptom was that
"Preset" trigger level was no longer correct, but the trigger level could
still be successfully set manually. For these cases, as the years progressed
(and the "rot" worsened), you had to turn the manual trigger level pot
further and further to one direction to find trigger. As the years of
deterioration continued, there came a point where both trigger level
setability and sensitivity were no longer acceptable.
If you have only a small trigger level issue in "Preset" mode, my thinking
is that you can still correct for it by doing the A3/A4 input adjustments.
Best,
Greg
----- Original Message -----
From: "Joseph Gray" jgray@zianet.com
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Saturday, October 10, 2009 2:23 PM
Subject: Re: [time-nuts] 5370A
Greg,
Yes, the trigger levels are near zero when the LEDs are flashing. The
level pots are both at 1 o'clock, instead of 12 o'clock, however.
There does seem to be a problem with the preset trigger postition.
Although the display shows 0.00V when at preset, I don't think it is.
When I manually set the levels to 0.00V, the TI shows 100nS. At preset
levels, the TI is noticeably higher.
Joe Gray
KA5ZEC
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.